Precise and high-throughput origin discrimination for green coffee beans by mass spectrometry-based metabolic
Zifan Yang1, Yuchen Feng1, Zihang Yang1
1School of Chemistry and Molecular Engineering, East China Normal University, Shanghai 200241, PR China.
Food Research International (Ottawa, Ont.)
|March 27, 2026
Summary
This study introduces a rapid, high-throughput method using nanoparticle-enhanced mass spectrometry and machine learning to accurately identify the origin of green coffee beans, combating food fraud.
Area of Science:
- Analytical Chemistry
- Food Science
- Biotechnology
Background:
- Falsely labeled origins of plant foods like green coffee beans pose a significant trade risk.
- Current origin discrimination tools lack precision and high-throughput capabilities.
Purpose of the Study:
- To develop a precise and high-throughput method for green coffee bean origin discrimination.
- To combat fraud in the trade of high-value plant foods.
Main Methods:
- Employed high-performance ferric nanoparticle-enhanced laser desorption/ionization mass spectrometry (NP-EDL-MS) for phytochemical analysis.
- Acquired metabolic fingerprints (approx. 500 features) of single green coffee beans within 30 seconds.
- Evaluated machine learning algorithms (Logistic Regression, SVM, Random Forest, KNN) for classification.
Main Results:
- Achieved high-throughput and high-sensitivity phytochemical analysis.
- Logistic Regression model reached 0.978 accuracy for multiclass origin classification (Gesha beans).
- Achieved an Area Under the Curve (AUC) of 0.996 for binary classification of high-value beans.
Conclusions:
- Developed a precise, high-throughput, and cost-effective approach for green coffee bean origin discrimination.
- Simplified classifiers using feature panels demonstrated strong validation performance.
- The method shows promise for broader applications in plant food origin verification.
